Face Recognition: Challenges and Issues in Smart City/Environments

Face Recognition: Challenges and Issues in Smart City/Environments

Table of Contents


Smart city challenges, like increased traffic, the risk to public safety, effective law enforcement, and the smart environment challenges improving personalized services such as health care and home environment need personal identification. Face recognition has proved to be useful and amicable bio-metric for smart city and smart environment challenges. In this paper, we review the use cases pertaining to smart city and smart environment leading to use case-specific requirements on Face recognition. We describe the open challenges in anti-spoofing and standardization to make face recognition of a foolproof system. For applications with demand for low power, we show that with proper considerations on training data, cost function, and model architecture we could build a low-complex CNN model with reasonable accuracy (91.4% on LFW data set).

  • Author Keywords

    • Face recognition,
    • smart city,
    • smart environment,
    • smart interaction
  • IEEE Keywords

    • Face recognition,
    • Face,
    • Complexity theory,
    • Smart cities,
    • Standardization,
    • Law enforcement,
    • Computational modeling


Rapid urbanization brought lot of challenges [1] to public life like increased traffic, the risk to public safety, and effective law enforcement. We can call them smart city challenges. At the same time with intelligent computing becoming ubiquitous [2], every device that humans interact with becoming smarter day by day. This brings up questions on how to personalize services like health care, home environment, etc. In both the smart city challenges and smart environment or smart interaction challenges, recognizing people is a must. Of the wide range of biometrics like fingerprint recognition, iris recognition, face recognition, etc., face recognition (FR) is favored the most for the reasons like a) FR is the natural way of recognizing people, b) is non-invasive, and c) low cost. For the same reasons, the demand for FR has grown rapidly in both smart cities and smart environments use cases. In fact, a recent market study [3] predicted that FR system will have a market worth of $7.0 billion by 2024. In this paper, we review the use cases and attempt to summarize the FR requirements and open problems. Particularly, in Section II, we review the use cases and categorize the requirements. In Section III, we present our approach on how to trade-off the complexity and accuracy of an FR system. In Sections IV and V, we talk about the open problems in face anti-spoofing and FR system standardization, respectively. Finally, in Section VI, we summarize the paper with future directions.

Use Cases And Requirements

In this section, first we briefly describe different use cases that would need face recognition, then we attempt to categorize the use cases into different classes based on the type of requirements they would impose on face recognition.

  1. FR use cases

We categorize the use cases into four broad classes and list some of them under each class. Note that this is not an exhaustive review, but acts as a reference to understand the different requirements on face recognition. We have also tagged the use cases proposed by us.

  • Health Care: Hospitals and elderly care homes are investing heavily on face recognition to monitor sick patients. Some of the aspects covered are as follows.
  • Patient check-in process
  • Patient tracking
  • Managing medicine distribution
  • Diagnosis. Ex: Face2gene [4] uses face recognition to diagnose certain disease
  • Staff identification
  • Retrieve critical information of patients in an emergency. Ex: In case of accidents, help public (or), ambulance drivers, to identify the blood group of the victim and their details [proposed]

2) Smart Homes and Smart Rooms: Use face recognition to ensure maximum security and to customize user-specific services. Examples:


  • Smart doorbell with face recognition- to identify any entrants
  • Customization in the travel industry as well as smart rooms using face recognition and identifying unique users to personalize surroundings to them
  • Person-specific recommendation systems and services for smart homes
  • Automatic floor selection in Elevators for aged and disabled people like blind [proposed]

3) Security, Access Control & Law Enforcement:

  • Easy boarding and security checks at airports
  • Jaywalking violations, as used in China [5]
  • To book traffic violations by rental transport users (such as bounce, vogo, drivezy) [proposed]
  • Identifying victims of human trafficking (E.g. to identify missing children)[proposed]
  • Criminal identification in the context of the environment [proposed]
  • Identification of men in female-reserved coaches (or) women-only areas like hostels [proposed]
  • To build a unified penalty system (FR-based fine collection system for all kind of violations: traffic police, public transport usage without a valid ticket) [proposed]

4) Marketing & Retail:

  • Payments on public transports (E.g. metro trains, public buses, toll gates, electricity bill payments)
  • Face recognition for retail stores to make shopping more convenient
  • Face recognition for supermarkets to track shoppers with suspicious behavior and identify shoplifters • Identify users based on face recognition for customized advertisements and promotions

Future Directions

In this review, we looked at different use cases and their requirements on the FR system. Then, we briefly discussed the challenges in anti-spoofing and the need for standardization. We have also proposed a low-complex model that can be used in small-scale real-time applications. In summary, to develop and deploy FR systems for a wide range use cases, following would be the broad list of open problems to be addressed: • Development of NIR based scalable and robust FR models • Development of FR models that are scalable yet low complex for large-scale deployment • Development of FR systems that are robust against a variety of fake face attacks • Effective and comprehensive standardization of FR systems addressing ethical, privacy, and bias and decision-making issues.

About KSRA

The Kavian Scientific Research Association (KSRA) is a non-profit research organization to provide research / educational services in December 2013. The members of the community had formed a virtual group on the Viber social network. The core of the Kavian Scientific Association was formed with these members as founders. These individuals, led by Professor Siavosh Kaviani, decided to launch a scientific / research association with an emphasis on education.

KSRA research association, as a non-profit research firm, is committed to providing research services in the field of knowledge. The main beneficiaries of this association are public or private knowledge-based companies, students, researchers, researchers, professors, universities, and industrial and semi-industrial centers around the world.

Our main services Based on Education for all Spectrum people in the world. We want to make an integration between researches and educations. We believe education is the main right of Human beings. So our services should be concentrated on inclusive education.

The KSRA team partners with local under-served communities around the world to improve the access to and quality of knowledge based on education, amplify and augment learning programs where they exist, and create new opportunities for e-learning where traditional education systems are lacking or non-existent.

FULL Paper PDF file:

Face Recognition: Challenges and Issues in Smart City/Environments



G. B. Praveen and J. Dakala, “Face Recognition




Challenges and Issues in Smart City/Environments

Publish in

2020 International Conference on COMmunication Systems & NETworkS (COMSNETS), Bengaluru, India, 2020, pp. 791-793



PDF reference and original file: Click here

+ posts

Somayeh Nosrati was born in 1982 in Tehran. She holds a Master's degree in artificial intelligence from Khatam University of Tehran.

Website | + posts

Professor Siavosh Kaviani was born in 1961 in Tehran. He had a professorship. He holds a Ph.D. in Software Engineering from the QL University of Software Development Methodology and an honorary Ph.D. from the University of Chelsea.

Website | + posts

Nasim Gazerani was born in 1983 in Arak. She holds a Master's degree in Software Engineering from UM University of Malaysia.